Face Generation¶

In this project, you'll use generative adversarial networks to generate new images of faces.

Get the Data¶

You'll be using two datasets in this project:

MNIST
CelebA

Since the celebA dataset is complex and you're doing GANs in a project for the first time, we want you to test your neural network on MNIST before CelebA. Running the GANs on MNIST will allow you to see how well your model trains sooner.

If you're using FloydHub, set data_dir to "/input" and use the FloydHub data ID "R5KrjnANiKVhLWAkpXhNBe".

data_dir = './data'

# FloydHub - Use with data ID "R5KrjnANiKVhLWAkpXhNBe"
#data_dir = '/input'


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper

helper.download_extract('mnist', data_dir)
helper.download_extract('celeba', data_dir)

Found mnist Data
Found celeba Data

Explore the Data¶

MNIST¶

As you're aware, the MNIST dataset contains images of handwritten digits. You can view the first number of examples by changing show_n_images.

show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
%matplotlib inline
import os
from glob import glob
from matplotlib import pyplot

mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'mnist/*.jpg'))[:show_n_images], 28, 28, 'L')
pyplot.imshow(helper.images_square_grid(mnist_images, 'L'), cmap='gray')

<matplotlib.image.AxesImage at 0x7fdd1fa1d390>

CelebA¶

The CelebFaces Attributes Dataset (CelebA) dataset contains over 200,000 celebrity images with annotations. Since you're going to be generating faces, you won't need the annotations. You can view the first number of examples by changing show_n_images.

show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'img_align_celeba/*.jpg'))[:show_n_images], 28, 28, 'RGB')
pyplot.imshow(helper.images_square_grid(mnist_images, 'RGB'))

<matplotlib.image.AxesImage at 0x7fdd1f944f98>

Preprocess the Data¶

Since the project's main focus is on building the GANs, we'll preprocess the data for you. The values of the MNIST and CelebA dataset will be in the range of -0.5 to 0.5 of 28x28 dimensional images. The CelebA images will be cropped to remove parts of the image that don't include a face, then resized down to 28x28.

The MNIST images are black and white images with a single color channel while the CelebA images have 3 color channels (RGB color channel).

Build the Neural Network¶

You'll build the components necessary to build a GANs by implementing the following functions below:

model_inputs
discriminator
generator
model_loss
model_opt
train

Check the Version of TensorFlow and Access to GPU¶

This will check to make sure you have the correct version of TensorFlow and access to a GPU

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf

# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer.  You are using {}'.format(tf.__version__)
print('TensorFlow Version: {}'.format(tf.__version__))

# Check for a GPU
if not tf.test.gpu_device_name():
    warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
    print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))

TensorFlow Version: 1.1.0
Default GPU Device: /gpu:0

Input¶

Implement the model_inputs function to create TF Placeholders for the Neural Network. It should create the following placeholders:

Real input images placeholder with rank 4 using image_width, image_height, and image_channels.
Z input placeholder with rank 2 using z_dim.
Learning rate placeholder with rank 0.

Return the placeholders in the following the tuple (tensor of real input images, tensor of z data)

import problem_unittests as tests

def model_inputs(image_width, image_height, image_channels, z_dim):
    """
    Create the model inputs
    :param image_width: The input image width
    :param image_height: The input image height
    :param image_channels: The number of image channels
    :param z_dim: The dimension of Z
    :return: Tuple of (tensor of real input images, tensor of z data, learning rate)
    """
    # TODO: Implement Function
    input_real = tf.placeholder(tf.float32, shape=(None, image_width, image_height, image_channels), name="input_real")
    input_z = tf.placeholder(tf.float32, shape=(None, z_dim), name='input_z')
    learning_rate = tf.placeholder(tf.float32, name="learning_rate")

    return input_real, input_z, learning_rate


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_inputs(model_inputs)

Tests Passed

Discriminator¶

Implement discriminator to create a discriminator neural network that discriminates on images. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "discriminator" to allow the variables to be reused. The function should return a tuple of (tensor output of the discriminator, tensor logits of the discriminator).

def discriminator(images, reuse=False, alpha=0.2):
    """
    Create the discriminator network
    :param images: Tensor of input image(s)
    :param reuse: Boolean if the weights should be reused
    :return: Tuple of (tensor output of the discriminator, tensor logits of the discriminator)
    """
    # TODO: Implement Function
   
    with tf.variable_scope("discriminator", reuse=reuse):
        
        x1 = tf.layers.conv2d(images, 64, 5, strides=2, padding='same')
        relu1 = tf.maximum(alpha * x1, x1)
        
        x2 = tf.layers.conv2d(relu1, 128, 5, strides=2, padding='same')
        bn2 = tf.layers.batch_normalization(x2, training=True)
        relu2 = tf.maximum(alpha * bn2, bn2)
        
        x3 = tf.layers.conv2d(relu2, 256, 5, strides=2, padding='same')
        bn3 = tf.layers.batch_normalization(x3, training=True)
        relu3 = tf.maximum(alpha * bn3, bn3)
        
        flat = tf.reshape(relu3, (-1, 4*4*256))
        logits = tf.layers.dense(flat, 1)
        out = tf.sigmoid(logits)
        

    return out, logits


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_discriminator(discriminator, tf)

Tests Passed

Generator¶

Implement generator to generate an image using z. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "generator" to allow the variables to be reused. The function should return the generated 28 x 28 x out_channel_dim images.

def generator(z, out_channel_dim, is_train=True):
    """
    Create the generator network
    :param z: Input z
    :param out_channel_dim: The number of channels in the output image
    :param is_train: Boolean if generator is being used for training
    :return: The tensor output of the generator
    """
    # TODO: Implement Function
    with tf.variable_scope("generator", reuse=not is_train):
        
        alpha=0.2 
        
        fc1 = tf.layers.dense(z, 7*7*512)
        
        
        x1 = tf.reshape(fc1, (-1, 7, 7, 512))
        x1 = tf.layers.batch_normalization(x1, training=is_train)
        x1 = tf.maximum(alpha * x1, x1)
        print (x1)
        
        x2 = tf.layers.conv2d_transpose(x1, 256, 1, strides=1, padding='SAME')
        x2 = tf.layers.batch_normalization (x2, training=is_train)
        x2 = tf.maximum(alpha * x2, x2)
        print (x2)
        
        x3 = tf.layers.conv2d_transpose(x2, 128, 3, strides=2, padding ='SAME')
        x3 = tf.layers.batch_normalization(x3, training=is_train)
        x3 = tf.maximum(alpha * x3, x3)
        print (x3)
        
        logits = tf.layers.conv2d_transpose(x3, out_channel_dim, 3, strides=2, padding='SAME')
        
        out = tf.tanh(logits)
        print (logits)
        # expect 28
    
    return out


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_generator(generator, tf)

Tensor("Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 5), dtype=float32)
Tensor("Maximum_3:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("Maximum_4:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("Maximum_5:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("conv2d_transpose_6/BiasAdd:0", shape=(?, 28, 28, 5), dtype=float32)
Tests Passed

Loss¶

Implement model_loss to build the GANs for training and calculate the loss. The function should return a tuple of (discriminator loss, generator loss). Use the following functions you implemented:

discriminator(images, reuse=False)
generator(z, out_channel_dim, is_train=True)

def model_loss(input_real, input_z, out_channel_dim):
    """
    Get the loss for the discriminator and generator
    :param input_real: Images from the real dataset
    :param input_z: Z input
    :param out_channel_dim: The number of channels in the output image
    :return: A tuple of (discriminator loss, generator loss)
    """
    # TODO: Implement Function
    alpha = 0.2
    g_model = generator(input_z, out_channel_dim)
    d_model_real, d_logits_real = discriminator(input_real)
    d_model_fake, d_logits_fake = discriminator(g_model, reuse=True)
    
    d_loss_real = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_real, labels=tf.ones_like(d_model_real)*(1-0.1)))
    d_loss_fake = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.zeros_like(d_model_fake)))
    g_loss = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.ones_like(d_model_fake)))
    
    d_loss = d_loss_real + d_loss_fake
    return d_loss, g_loss


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_loss(model_loss)

Tensor("generator/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 4), dtype=float32)
Tests Passed

Optimization¶

Implement model_opt to create the optimization operations for the GANs. Use tf.trainable_variables to get all the trainable variables. Filter the variables with names that are in the discriminator and generator scope names. The function should return a tuple of (discriminator training operation, generator training operation).

def model_opt(d_loss, g_loss, learning_rate, beta1):
    """
    Get optimization operations
    :param d_loss: Discriminator loss Tensor
    :param g_loss: Generator loss Tensor
    :param learning_rate: Learning Rate Placeholder
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :return: A tuple of (discriminator training operation, generator training operation)
    """
    # TODO: Implement Function
    
    t_vars = tf.trainable_variables()
    d_vars = [var for var in t_vars if var.name.startswith('discriminator')]
    g_vars = [var for var in t_vars if var.name.startswith('generator')]

    # Optimize
    with tf.control_dependencies(tf.get_collection(tf.GraphKeys.UPDATE_OPS)):
        d_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(d_loss, var_list=d_vars)
        g_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(g_loss, var_list=g_vars)

    return d_train_opt, g_train_opt


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_opt(model_opt, tf)

Tests Passed

Neural Network Training¶

Show Output¶

Use this function to show the current output of the generator during training. It will help you determine how well the GANs is training.

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np

def show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
    """
    Show example output for the generator
    :param sess: TensorFlow session
    :param n_images: Number of Images to display
    :param input_z: Input Z Tensor
    :param out_channel_dim: The number of channels in the output image
    :param image_mode: The mode to use for images ("RGB" or "L")
    """
    cmap = None if image_mode == 'RGB' else 'gray'
    z_dim = input_z.get_shape().as_list()[-1]
    example_z = np.random.uniform(-1, 1, size=[n_images, z_dim])

    samples = sess.run(
        generator(input_z, out_channel_dim, False),
        feed_dict={input_z: example_z})

    images_grid = helper.images_square_grid(samples, image_mode)
    pyplot.imshow(images_grid, cmap=cmap)
    pyplot.show()

Train¶

Implement train to build and train the GANs. Use the following functions you implemented:

model_inputs(image_width, image_height, image_channels, z_dim)
model_loss(input_real, input_z, out_channel_dim)
model_opt(d_loss, g_loss, learning_rate, beta1)

Use the show_generator_output to show generator output while you train. Running show_generator_output for every batch will drastically increase training time and increase the size of the notebook. It's recommended to print the generator output every 100 batches.

def train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode):
    """
    Train the GAN
    :param epoch_count: Number of epochs
    :param batch_size: Batch Size
    :param z_dim: Z dimension
    :param learning_rate: Learning Rate
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :param get_batches: Function to get batches
    :param data_shape: Shape of the data
    :param data_image_mode: The image mode to use for images ("RGB" or "L")
    """
    # TODO: Build Model
    
    steps = 0
    samples, losses = [], []
    
    input_real, input_z, lr = model_inputs(data_shape[1], data_shape[2], data_shape[3], z_dim)
    d_loss, g_loss = model_loss(input_real, input_z, data_shape[3])
    d_train_opt, g_train_opt = model_opt(d_loss, g_loss, lr, beta1)
    
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())
        for epoch_i in range(epoch_count):
            for batch_images in get_batches(batch_size):
                # TODO: Train Model
                steps += 1

                # Sample random noise for G
                batch_z = np.random.uniform(-1, 1, size=(batch_size, z_dim))

                # Run optimizers
                _ = sess.run(d_train_opt, feed_dict={input_real: batch_images, input_z: batch_z, lr: learning_rate})
                #_ = sess.run(g_train_opt, feed_dict={input_z: batch_z, input_real: x, lr: learning_rate})
                _ = sess.run(g_train_opt, feed_dict={input_real: batch_images, input_z: batch_z, lr: learning_rate})

                if steps % 50 == 0:
                    # At the end of each epoch, get the losses and print them out
                    train_loss_d = d_loss.eval({input_z: batch_z, input_real: batch_images})
                    train_loss_g = g_loss.eval({input_z: batch_z})

                    print("Epoch {}/{}...".format(epoch_i+1, epochs),
                          "Discriminator Loss: {:.4f}...".format(train_loss_d),
                          "Generator Loss: {:.4f}".format(train_loss_g))
                    # Save losses to view after training
                    losses.append((train_loss_d, train_loss_g))

                if steps % 100 == 0:
                    sample_z = np.random.uniform(-1, 1, size=(16, z_dim))
                    gen_samples = sess.run(
                                   generator(input_z, data_shape[3], is_train=False),
                                   feed_dict={input_z: sample_z})
                    samples.append(gen_samples)
                    # show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
                    show_generator_output(sess, 5, tf.convert_to_tensor(input_z), data_shape[3], data_image_mode)
                    #_ = view_samples(-1, samples, 6, 12, figsize=figsize)
                    #plt.show()

        #saver.save(sess, './checkpoints/generator.ckpt')

    #with open('samples.pkl', 'wb') as f:
    #    pkl.dump(samples, f)
    
    return losses, samples

MNIST¶

Test your GANs architecture on MNIST. After 2 epochs, the GANs should be able to generate images that look like handwritten digits. Make sure the loss of the generator is lower than the loss of the discriminator or close to 0.

batch_size = 16  #Try reducing batch_size to (~16 to 32)
z_dim = 100
learning_rate = 0.0005  #DCGAN with this architectural structure remains stable with lr between 0.0001 and 0.0008
beta1 = 0.5  #The beta1 is too high. Values like 0.3 to 0.5 have shown to get best results.


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 2

mnist_dataset = helper.Dataset('mnist', glob(os.path.join(data_dir, 'mnist/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, mnist_dataset.get_batches,
          mnist_dataset.shape, mnist_dataset.image_mode)

Tensor("generator/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)
Epoch 1/2... Discriminator Loss: 0.8019... Generator Loss: 2.1073
Epoch 1/2... Discriminator Loss: 1.5388... Generator Loss: 0.4260
Tensor("generator_1/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_1/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_1/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_1/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)
Tensor("generator_2/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_2/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_2/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_2/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)

Epoch 1/2... Discriminator Loss: 0.5035... Generator Loss: 2.2739
Epoch 1/2... Discriminator Loss: 0.5197... Generator Loss: 2.3893
Tensor("generator_3/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_3/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_3/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_3/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)
Tensor("generator_4/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_4/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_4/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_4/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)

Epoch 1/2... Discriminator Loss: 1.0708... Generator Loss: 1.2742
Epoch 1/2... Discriminator Loss: 1.8766... Generator Loss: 0.3306
Tensor("generator_5/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_5/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_5/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_5/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)
Tensor("generator_6/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_6/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_6/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_6/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)

Epoch 1/2... Discriminator Loss: 1.2117... Generator Loss: 0.6639
Epoch 1/2... Discriminator Loss: 0.6718... Generator Loss: 1.5278
Tensor("generator_7/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_7/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_7/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_7/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)
Tensor("generator_8/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_8/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_8/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_8/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)

Epoch 1/2... Discriminator Loss: 0.5671... Generator Loss: 1.7343
Epoch 1/2... Discriminator Loss: 1.1977... Generator Loss: 0.6624
Tensor("generator_9/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_9/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_9/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_9/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)
Tensor("generator_10/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_10/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_10/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_10/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)

Epoch 1/2... Discriminator Loss: 1.1203... Generator Loss: 0.6831
Epoch 1/2... Discriminator Loss: 0.9382... Generator Loss: 0.9745
Tensor("generator_11/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_11/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_11/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_11/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)
Tensor("generator_12/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_12/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_12/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_12/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)

Epoch 1/2... Discriminator Loss: 0.5101... Generator Loss: 1.9681
Epoch 1/2... Discriminator Loss: 0.6156... Generator Loss: 1.5568
Tensor("generator_13/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_13/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_13/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_13/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)
Tensor("generator_14/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_14/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_14/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_14/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)

Epoch 1/2... Discriminator Loss: 0.4386... Generator Loss: 2.6910
Epoch 1/2... Discriminator Loss: 1.1301... Generator Loss: 0.7679
Tensor("generator_15/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_15/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_15/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_15/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)
Tensor("generator_16/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_16/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_16/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_16/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)

Epoch 1/2... Discriminator Loss: 0.5747... Generator Loss: 2.4929
Epoch 1/2... Discriminator Loss: 0.6696... Generator Loss: 1.4412
Tensor("generator_17/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_17/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_17/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_17/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)
Tensor("generator_18/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_18/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_18/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_18/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)

Epoch 1/2... Discriminator Loss: 0.4136... Generator Loss: 2.7055
Epoch 1/2... Discriminator Loss: 0.8394... Generator Loss: 1.0225
Tensor("generator_19/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_19/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_19/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_19/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)
Tensor("generator_20/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_20/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_20/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_20/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 1), dtype=float32)

CelebA¶

Run your GANs on CelebA. It will take around 20 minutes on the average GPU to run one epoch. You can run the whole epoch or stop when it starts to generate realistic faces.

batch_size = 16  #Try reducing batch_size to (~16 to 32)
z_dim = 100
learning_rate = 0.0005  #DCGAN with this architectural structure remains stable with lr between 0.0001 and 0.0008
beta1 = 0.5  #The beta1 is too high. Values like 0.3 to 0.5 have shown to get best results.


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 1

celeba_dataset = helper.Dataset('celeba', glob(os.path.join(data_dir, 'img_align_celeba/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, celeba_dataset.get_batches,
          celeba_dataset.shape, celeba_dataset.image_mode)

Tensor("generator/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)
Epoch 1/1... Discriminator Loss: 2.5810... Generator Loss: 0.1562
Epoch 1/1... Discriminator Loss: 0.8485... Generator Loss: 4.1541
Tensor("generator_1/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_1/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_1/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_1/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)
Tensor("generator_2/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_2/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_2/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_2/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)

Epoch 1/1... Discriminator Loss: 1.9449... Generator Loss: 0.3195
Epoch 1/1... Discriminator Loss: 1.0545... Generator Loss: 0.8427
Tensor("generator_3/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_3/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_3/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_3/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)
Tensor("generator_4/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_4/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_4/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_4/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)

Epoch 1/1... Discriminator Loss: 1.8255... Generator Loss: 0.2982
Epoch 1/1... Discriminator Loss: 0.8210... Generator Loss: 1.3616
Tensor("generator_5/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_5/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_5/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_5/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)
Tensor("generator_6/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_6/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_6/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_6/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)

Epoch 1/1... Discriminator Loss: 1.7993... Generator Loss: 0.3590
Epoch 1/1... Discriminator Loss: 0.5474... Generator Loss: 1.9443
Tensor("generator_7/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_7/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_7/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_7/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)
Tensor("generator_8/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_8/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_8/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_8/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)

Epoch 1/1... Discriminator Loss: 0.3985... Generator Loss: 4.0781
Epoch 1/1... Discriminator Loss: 1.4472... Generator Loss: 7.6636
Tensor("generator_9/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_9/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_9/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_9/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)
Tensor("generator_10/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_10/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_10/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_10/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)

Epoch 1/1... Discriminator Loss: 0.5082... Generator Loss: 2.5054
Epoch 1/1... Discriminator Loss: 0.3818... Generator Loss: 3.4556
Tensor("generator_11/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_11/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_11/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_11/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)
Tensor("generator_12/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_12/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_12/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_12/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)

Epoch 1/1... Discriminator Loss: 0.4175... Generator Loss: 4.7357
Epoch 1/1... Discriminator Loss: 0.8411... Generator Loss: 1.5651
Tensor("generator_13/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_13/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_13/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_13/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)
Tensor("generator_14/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_14/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_14/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_14/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)

Epoch 1/1... Discriminator Loss: 1.0884... Generator Loss: 0.7400
Epoch 1/1... Discriminator Loss: 0.4014... Generator Loss: 2.8814
Tensor("generator_15/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_15/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_15/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_15/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)
Tensor("generator_16/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_16/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_16/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_16/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)

Epoch 1/1... Discriminator Loss: 0.3471... Generator Loss: 5.6436
Epoch 1/1... Discriminator Loss: 0.6552... Generator Loss: 1.5839
Tensor("generator_17/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_17/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_17/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_17/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)
Tensor("generator_18/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_18/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_18/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_18/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)

Epoch 1/1... Discriminator Loss: 0.3775... Generator Loss: 5.6968
Epoch 1/1... Discriminator Loss: 0.3745... Generator Loss: 4.3571
Tensor("generator_19/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_19/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_19/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_19/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)
Tensor("generator_20/Maximum:0", shape=(?, 7, 7, 512), dtype=float32)
Tensor("generator_20/Maximum_1:0", shape=(?, 7, 7, 256), dtype=float32)
Tensor("generator_20/Maximum_2:0", shape=(?, 14, 14, 128), dtype=float32)
Tensor("generator_20/conv2d_transpose_3/BiasAdd:0", shape=(?, 28, 28, 3), dtype=float32)

Submitting This Project¶

When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_face_generation.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "helper.py" and "problem_unittests.py" files in your submission.